Safety and Risk Analysis of Autonomous Vehicles Using Computer Vision and Neural Networks

Recursive Least Squares Based Refinement Network for Vehicle Trajectory Prediction

Trajectory prediction of surrounding objects plays a pivotal role in the field of autonomous driving vehicles. In the current rollout process, it suffers from an accumulation of errors, which has a negative impact on prediction accuracy. This paper proposes a parametric-learning recursive least-squares (RLS) method integrated with an encoder–decoder framework for trajectory prediction, named the recursive least-squares-based refinement network (RRN). Through the generation of several anchors in the future trajectory, RRN can capture both local and global motion patterns. We conducted experiments on the prevalent NGSIM and INTERACTION datasets, which contain various scenarios such as highways, intersections and roundabouts. The promising results indicate that RRN could improve the performance of the rollout trajectory prediction effectively.

Understanding the Motivation and Satisfaction of Private Vehicle Users in an Eastern European Country Using Heterogeneity Analysis

Transport service provision in many urban areas is dominated by car users, resulting in several traffic externality issues (e.g., noise, pollution, accidents). This paper investigates the perception and satisfaction of private vehicle (PV) users, including micro-mobility users, during their commute by car in an Eastern European country context. The study used empirical data from a sample of 500 commuters in Budapest, Hungary, between October and November 2020. To achieve a deeper understanding of the motivation and explore the perception of PV users towards using sustainable transport services. For analysis in this study, descriptive statistics and segmentation techniques were applied. The key findings indicate that PV users can be attracted to using sustainable transport by designing the travel service quality to provide the level of service desired by customers. Moreover, the majority (73%) of PV commuters were satisfied or very satisfied with the quality attributes of the car service, assessed on a scale of 1 to 5; at the same time, PV users agreed that using public transport helps towards improving the environment and serves to reduce problems derived from traffic. In addition, various elements influence transport choice; for example, results from ordered logit models (OLMs) indicate that security, relaxation, flexibility and comfort are the main significant attributes influencing PV users’ overall satisfaction with cars. The results suggest the necessity for a segmentation technique in the analysis of travel attitudes and satisfaction aimed at reducing the frequency of existing car use to enhance sustainable transportation.

Emerging Trends and Knowledge Structures of Smart Urban Governance

The concept of smart cities peaked in 2015, bringing an increased influx of ‘smart’ devices in the form of the Internet of Things (IoT) and sensors in cities. As a result, interest in smart urban governance has become more prevalent in administrative, organisational, and political circles. This is sustained by both local and global demands for an increased contribution to the goals of sustainability through urban governance processes in response to climate change urgencies. Cities generate up to 70% of global emissions, and in light of societal pressures for more inclusivity and democratic processes, the need for sound urban governance is merited. Further knowledge on the theme of smart urban governance is required to better understand the trends and knowledge structures and better assist policy design. Therefore, this study was undertaken to understand and map the evolution of the concept of smart urban governance through a bibliometric analysis and science mapping techniques using VOSviewer. In total, 1897 articles were retrieved from the Web of Science database over 5 decades, from 1968 to 2021, and divided into three subperiods, namely 1978 to 2015, 2016 to 2019, and 2020 to early 2022. Results indicate that the overall emerging themes across the three periods highlight the need for citizen participation in urban policies, especially in relation to smart cities, and for sustained innovation for e-participation, e-governance, and policy frameworks. The results of this study can aid both researchers exploring the concept of urban governance and policy makers rendering more inclusive urban policies, especially those hosting technological and digital domains.